Incorporando contexto externo

Engenharia rápida com a API OpenAI

Fouad Trad

Machine Learning Engineer

A necessidade de contexto externo

  • Modelos de linguagem pré-treinados reconhecem o que foi visto no treino
  • Precisa fornecer mais contexto
  • Mais precisão e eficácia

Ícone representando um chatbot em um celular.

Engenharia rápida com a API OpenAI

Falta de informação nos LLMs

  • Corte de conhecimento
system_prompt = "Act as a financial expert that knows about the latest trends."

user_prompt = "What are the top financial trends in 2023?"

print(get_response(system_prompt, user_prompt))

I apologize for any inconvenience, but as of my last knowledge update in 
September 2021, I don't have information about financial trends in 2023.
Engenharia rápida com a API OpenAI

Falta de informação nos LLMs

  • Solicitação de informação não pública
system_prompt = "Act as a study buddy that helps me with my studies to succeed 
in exams."


user_prompt = "What is the name of my favorite instructor?"

print(get_response(system_prompt, user_prompt))

I don't have personal information about you, including the name of your 
favorite instructor.
Engenharia rápida com a API OpenAI

Como fornecer informação extra?

  • Exemplos de conversas anteriores
  • Prompt do sistema

Ícone mostrando duas caixas de diálogo para refletir conversação.

Engenharia rápida com a API OpenAI

Conversas de exemplo

  • Guie o modelo para responder perguntas específicas
response = client.chat.completions.create(
  model="gpt-3.5-turbo",
  messages=[{"role": "system",
             "content": "You are a customer service chatbot that responds to user queries in a gentle way"},

            {"role": "user",
             "content": "What services do you offer?"},

           {"role": "assistant",
             "content": "We provide services for web application development, mobile app development, and 
            custom software solutions."},

           {"role": "user",
             "content": "How many services do you have?"}])

print(response.choices[0].message.content)

We have 3 services including web application development, mobile app development, and custom software solutions.
  • Desvantagem: pode exigir muitas amostras
Engenharia rápida com a API OpenAI

Prompt do sistema

  • Inclui o contexto necessário para respostas do chatbot
services = "ABC Tech Solutions, a leading IT company, offers a range of services: 
application development, mobile app development, and custom software solutions."

system_prompt = f"""You are a customer service chatbot that responds to user 
queries in a gentle way. Some information about our services are delimited by 
triple backticks.
```{services}```"""

user_prompt = "How many services do you offer?"
print(get_response(system_prompt, user_prompt))

We have 3 services including web application development, mobile app development, 
and custom software solutions.
Engenharia rápida com a API OpenAI

Nota final

  • Métodos anteriores funcionam bem para contextos pequenos
  • Contextos maiores exigem técnicas mais sofisticadas

Ícone representando um alerta.

Engenharia rápida com a API OpenAI

Vamos praticar!

Engenharia rápida com a API OpenAI

Preparing Video For Download...